Nontypable Haemophilus influenzae is a common cause of localized respiratory tract disease, especially otitis media, sinusitis, pneumonia, and bronchitis. Following each episode of acute otitis media, fluid remains in the middle ear for weeks to months and is associated with significant hearing deficit, which in turn can impair language acquisition, speech development, and school performance. The initial step in the pathogenesis of disease due to nontypable H. influenzae involves colonization of the upper respiratory epithelium. We have demonstrated that two related high-molecular-weight proteins called HMW1 and HMW2 promote attachment to human epithelium, an essential step in the process of colonization. Of note, HMW1- and HMW2-1ike proteins are present in 75-80% of all nontypable H. influenzae strains. In recent work, we have established that the HMW1 and HMW2 proteins are variant members of the autotransporter family and are glycosylated. Based on studies with HMW1, glycosylation requires a protein called HMW1C and a phosphoglucomutase involved in LOS biosynthesis. In addition, we have discovered that expression of HMW1 and HMW2 is phase variable, enabling the organism to adapt to diverse environments and evade the host immune response. In the present proposal we plan to elucidate the molecular details of HMW1 and HMW2 glycosylation. In particular, we will define the chemical structure of the carbohydrate modifying HMW1 and HMW2, the biosynthetic pathway involved in glycosylation of HMW 1 and HMW2, and the relationship between glycosylation and immunogenicity. In additional experiments, we will elucidate the mechanism by which expression of HMW1 and HMW2 is regulated, focusing in particular on the function of 7-base pair tandem repeats that lie within the promoters of the HMW1 and HMW2 structural genes and undergo spontaneous variation in number. We will also examine the role of a conserved 19-base pair sequence that is upstream of the repeats in the promoters of the HMW1 and HMW2 structural genes and upstream of the hmwB and hmwC genes. Finally, using microarray technology, we will identify H. influenzae genes that are activated by HMW1/HMW2-mediated adherence, then assess the effect of these genes in the chinchilla otitis media model. From a practical perspective, the results of these studies may be directly relevant to the development of novel antimicrobials and a licensed vaccine effective in the treatment and prevention of nontypable H. influenzae disease. More generally, they may provide fundamental insights into host-microbial relationships, protein secretion, protein glycosylation, and gene regulation.